Introduction

As the old Chinese saying goes, “If one’s diet is not regulated, it can lead to numerous diseases.” Food and health issues have garnered significant global attention (Clark et al., 2019). The global diet is shifting towards patterns associated with degenerative diseases, particularly high-protein and high-fat intake (Popkin, 2004). Research showed that 26% of global preventable deaths are attributed to poor diet, surpassing other risk factors such as smoking (GBD, 2017). This trend has been accompanied by a rapid increase in national healthcare investment, with global spending reaching $7.9 trillion in 2020 (WHO, 2021). Undoubtedly, the doubling of global health expenditures has led to significant improvements, with average life expectancy rising to 73.3 years in 2019 (WHO, 2023). However, the risk of diet-related non-communicable diseases (NCDs) is rising, suggesting a vicious cycle involving diet, health, and healthcare investment (UN, 2022).

Recognizing that diet globally contributes to an increased risk of disease, research on the burden of disease associated with diet has begun to accumulate. One group of studies focused on the relationship between overweight, obesity and healthcare costs (Tremmel et al., 2017). Most research relies on Jones’s two-part model (2000) and the instrumental variable method to estimate healthcare expenditures (Flegal et al., 2010; Cawley and Meyerhoefer, 2012; Qin and Pan, 2016). Other studies have explored the relationship between individual diets and healthcare expenditures using data from longitudinal cohort experiments. Some studies utilized large cohort data with extensive sample sizes, while others used small, precise data with well-defined individual characteristics (Sander and Bergemann, 2003). These studies include statistical experimental research (Sichieri et al., 2007; Kang et al., 2011; Lin et al., 2019) and clinical medical experiments examining the relationship between dietary behavior and healthcare expenditures (Dall et al., 2009; Zhang et al., 2017; Losina et al., 2019;). Additionally, some studies have quantified the disease risk associated with dietary factors and estimated the medical costs related to these diseases (Chiuve et al., 2012; Schwingshackl and Hoffmann, 2015; Biswas et al., 2015). Dietary factors include trace elements like salt, sugar, iron, vitamin D, and cellulose (Abdullah et al., 2015; Wilson et al., 2016; Dötsch-Klerk et al., 2023), as well as specific food types such as grains, vegetables, fruits, legumes, dairy products, and meat (Barnard et al., 1995; Doidge et al., 2012; Abdullah et al., 2017; Ekwaru et al., 2017; Javanbakht et al., 2018). Recently, there has been increasing attention on the healthcare expenditures associated with dietary patterns, including specific diets such as the Mediterranean diet, suboptimal diets, and diets in countries like Canada and the United States (Abdullah et al., 2017; Jardim et al., 2019; Jones et al., 2019; Scrafford et al., 2019; Liu et al., 2020; Herman et al., 2021). Most studies indicated that the disease burden of dietary significantly impacts population healthcare expenditures (Farasat and Salis, 2016).

Although previous studies have extensively discussed healthcare expenditures related to micro-scale diet, focusing on dietary factors and diseases using statistical and experimental data, the impact of changes in the national dietary structure on macro-scale healthcare expenditures is also crucial and cannot be overlooked. Most studies examining national healthcare expenditures focus on factors like population aging and air quality (Tenand, 2016). However, rational dietary adjustments are also a key component of individual preventive measures. The evolution of dietary structures has led to indirect costs of obesity in China amounting to US $4.3 billion (6.3% of GDP) (Popkin et al., 2006). Given the dietary shift towards degenerative diseases, further research is needed to understand the relationship between dietary changes and national health expenditures, as well as their impact on the National Health Service. Additionally, exploring the effects of dietary changes on medical expenditures is essential for optimizing costs, reducing the financial burden, and promoting healthy eating.

Therefore, this study analyzed changes in the dietary structure of Chinese residents from 1992 to 2021, focusing on macro-nutrients and food diversity, using data on food consumption and national healthcare expenditure. Additionally, the study investigates examines the evolution of healthcare expenditure within the national economic system. On this basis, we employed a fixed-effect model to explore how changes in dietary structure affect medical expenditure. Then, we used the peer pressure method and the gray prediction model to forecast future changes in China’s diet and medical expenditure. This approach aims to align national healthcare spending with the population’s health needs, thereby supporting the goals of “Healthy China 2030” and the development of a global community of health.

Materials and methods

General thought

Food consumption affects planetary health and is directly linked to the Sustainable Development Goals. Unsustainable diets represent a key factor in the nutrition–health–environment triad dilemma (Wang et al., 2022; Dou and Liu, 2024). Both the food system and healthcare system impact human health. Specifically, in the food system, dietary imbalances can lead to malnutrition, overnutrition, hidden hunger, and various health issues (Wang et al., 2023; Shang et al., 2023). Furthermore, food consumption induces a range of environmental impacts, including air pollution, that exacerbates the mortality rates of diet-related diseases (Springmann et al., 2023; Zhu et al., 2023a; Zhu et al., 2023b; Rampalli et al., 2023). Within the healthcare system, inputs in various aspects like financial allocations play a pivotal role in shaping public health (Farasat et al., 2016). In times of illness, medical care acts as the primary means to manage health risks, serving as a therapeutic investment in enhancing well-being (Grossman, 1972). Enhancing dietary patterns and healthcare allocations are pivotal tools to advance population health.

In addition, Additionally, the food and medical systems are interconnected through the population’s health. An increased risk of disease resulting from improper diets among residents often leads individuals to seek medical treatment and adopt other health measures. This increases personal and governmental expenditure on medical care. Furthermore, the rise in patient numbers due to improper diets leads to higher investments in medical personnel, hospital construction, and related areas, which are financed by the state and government. These processes create a negative feedback loop where poor eating behaviors lead to excessive consumption of medical resources. Conversely, increased medical investment due to eating behaviors stabilizes the quality of healthcare, thereby enhancing residents’ health. This also creates a positive feedback loop, where improved health leads to greater investment in health risk prevention and control. Food and healthcare systems are characterized by the interaction between diet and healthcare expenditures, which influence each other.

Disease risk is influenced by diet, medical expenditures are influenced by disease prevalence, and diet is influenced by healthcare expenditures. Food, health, and medical care form a dynamic circular system that interacts and balances within the framework of food and medical systems (Fig. 1). However, many regions experience a phenomenon where increased healthcare investment leads to more hospitals and higher patient numbers, which appears to result from an imbalance between preventive and treatment expenditures. Moreover, the tilting of inputs from governmental and social expenditures has not brought about a desirable reduction in individual cash expenditures. The orientation of human activities toward health is both a prerequisite and a continuous driver of socio-economic development (Dutton et al., 2018). The impact of dietary restructuring on national healthcare expenditures should be investigated.

Fig. 1
figure 1

The logic framework of this study.

Full size image

Analysis of Dietary Structure Evolution

Dietary diversity level

Dietary structure pertains to the proportion of various food types consumed. This study, based on GBD 2017 results, classifies food into eight categories: grains (including cereals, beans, and potatoes), vegetables, fruits, oilseeds, eggs, dairy products, meat, and aquatic products. Dietary diversity indicates the variety in overall food intake. Economic literature has noted that even limited food consumption can impact dietary diversity (Kant et al., 1993). Consequently, the concept of dysprosium (E) is used to quantify dietary diversification, providing a clear reflection of dietary structure. Food consumption data were sourced from the China Statistical Yearbook and the China Rural Statistical Yearbook for the period from 1992 to 2021. The calculation formula is as follows:

$${{DDI}}^{{country},{Urban},{Rural}}=\mathop{\sum }\limits_{i=1}^{n}{f}_{i}\mathrm{ln}\left(1/{f}_{i}\right)$$
(1)

Where, DDI takes the value of (0,1), the larger the value, the higher the degree of diversification of the dietary structure. n represents the number of food types and fi represents the proportion of annual per capita food consumption.

Dietary nutrient intake

Additionally, we introduced the concept of nutrient intake, including per capita daily intakes of energy, protein, fat, dietary fiber, and carbohydrates, to analyze the characteristics of the Chinese dietary structure. We used the balanced diet outlined in the Chinese Dietary Guidelines 2022 (CNS, 2022) as a model for nutritious diets to elucidate changes in diet quality. Food nutrient data were sourced from the Chinese Food Composition Table (https://nlc.chinanutri.cn/fq/), while the balanced dietary pattern was based on the recommended values from the Chinese Dietary Guidelines. Tables S1 and S2 in the Supplementary Materials provide data on food nutrient components and the balanced dietary pattern. The calculation formula is as follows:

$${{Nutrient}}_{{Urban}\,{\rm{or}}\,{Rural}}=\frac{\sum{n}_{i,j{{\times}}{f}_{i}}}{365}$$
(2)

Where, Nutrient refers to the daily intake of nutrients (kcal or g); fi is the per capita food consumption (kg/ person); ni,j is the representative value of the nutrient food.

Effects of dietary changes on health expenditure

Fixed effect model

To investigate the impact of dietary structure on healthcare expenditures, it is important to consider various factors, including the size of the economy and population. To better cover multiple levels of healthcare expenditure and deal with individual heterogeneity and endogeneity, this study employed a fixed effects model to analyze the impact of dietary structure on healthcare expenditures and other control variables across 31 provinces in China (excluding Hong Kong, Macao, and Taiwan) from 1992 to 2021. This model is useful for estimating causality in panel data, which is commonly used in economics and statistics, and is frequently applied in social science research (Antonakis et al., 2021). The data were obtained from the China Statistical Yearbook, statistical yearbooks of provinces (cities), China Price Survey Yearbook, China Urban (Town) Residents’ Household Living and Price Survey Yearbook and China Environmental Yearbook, and China Health and Health Statistics Yearbook (http://www.stats.gov.cn/).

For variable selection, food consumption was selected as the explanatory variable. In the statistics of China’s healthcare system, healthcare expenditures were categorized into government health expenditures, social health expenditures, and individual cash expenditures. Of these, government health expenditures refer to all levels of government for healthcare services, Medicare subsidies, health and Medicare administration, population and family planning expenditures, and other endeavors. Social health expenditures refer to the community’s financial contributions to health services, excluding government expenditures. Individual cash health expenditures are the payments residents make for various medical and health services. Total health costs represent the overall monetary amount of resources mobilized from society for health services in a country or region over a specific period. In this study, the indicators that represent healthcare expenditures in the provincial panel statistics are local financial health expenditures, household health expenditures, and per capita healthcare expenditures, including individual cash expenditures and total health costs in the statistical caliber of the health system.

Additionally, China’s demographic dividend is shifting from “quantity” to “quality” (Cai, 2022), with health emerging as a key component of “quality.” The impact of diet on human health can vary depending on the intake of food nutrients (He et al., 2021). Grossman’s health needs model assumes that healthcare expenditures are influenced by several factors, such as education, environmental, and economic development (Grossman, 1972). Factors such as the level of education (Sato, 2012), population aging (Ellis et al., 2013), the level of urbanization, and the level of income (Ross et al., 2007) have been shown to correlate with healthcare expenditures. Additionally, healthcare resources in the delivery system, such as the number of health service providers and beds, impact healthcare expenditures. Therefore, this study used control variables including years of education, urbanization rate, GDP per capita, disposable income per capita, aging rate, SO2 emissions, number of healthcare personnel, healthcare institutions, beds, health checkups, and mortality rate (Table S3). The model construction is as follows:

$${{ME}}_{i,t}={{{\beta }}}_{0}+{{{\beta }}}_{0}{f}_{1,t}+{{{\beta }}}_{2}{f}_{2,t}+\cdots +{{{\beta }}}_{j}{f}_{j,t}+\mathop{\sum }\limits_{q=1}^{n}{\delta }_{q}{C}_{i,t}+{{{\lambda }}}_{t}+{{{\mu }}}_{t}+{\varepsilon }_{i,t}$$
(3)

Where, MEit is the i th medical input index value in the t year, β0 is the intercept term, βjt (j = 1, 2,… 10) refers to the regression coefficient of dietary consumption fi,t. δq is the influence coefficient of each control variable, μi is the individual fixed effect, λt is the time fixed effect, εi,t is the random disturbance term.

Robustness test

We employed the control variable method and adjusted the sample size to test the model’s stability and verify its stationarity. First, we removed specific variables. The results of the fixed effects regression after removing the variable “years of education level” are presented in Table S4. For the core explanatory variables, the signs and significance of the coefficients for food consumption and nutrient intake on healthcare expenditure remained consistent, indicating the model’s robustness. Second, we adjusted the sample size. Medical expenditure was influenced by the COVID-19 pandemic during 2020–2021. We excluded the time series data from 2020 and 2021, using only panel data from 1992 to 2019 (Table S5) to reduce the impact of outliers on the regression results. The signs and significance of the coefficients for each explanatory variable on the outcome variables remained consistent, confirming the model’s robustness.

Prediction analysis of medical expenditure under the evolution of dietary structure

The peer pressure approach sets target values by referencing countries or regions with characteristics similar to China, such as income level. This study references Japan, South Korea, the United States, and Australia to simulate changes in dietary structure. We used GDP per capita as a reference to predict the Chinese dietary structure for 2025, 2030, 2035, 2040, 2045, and 2050. Specifically, China’s per capita GDP was $10,234 in 2021 and is projected to reach $14,590 by 2025 (Table S6). Japan’s per capita GDP reached this level in 1986 and 1987 (Table S7). Thus, we used the rate of change in food consumption in Japan during this period as a reference to estimate the rate of change in China’s food consumption. We also calculated the rate of change for other countries over different periods using the same method (Tables S8–S9). We averaged these values to determine the change rate of China’s food consumption, from which the Chinese dietary structure and nutrient intake were derived. Per capita GDP data were sourced from the World Bank (https://data.worldbank.org/indicator/NY.GDP.PCAP.CD). Projected GDP per capita figures for China are based on data published by the Research Group on Modernization in Chinese Style. Data on food consumption by country for various years were obtained from FAOSTAT (https://www.fao.org/faostat/en/#data/FBSH).

Additionally, we estimated changes in foodborne medical expenditures using the medical expenditure effect model. Moreover, we employed a gray model indicated by Deng (Deng, 1982), to forecast healthcare expenditures. This model enabled us to compare healthcare expenses related to dietary factors with the disparity in healthcare expenditure changes at different levels. The calculation formula is as follows:

$${\hat{X}}^{\left(0\right)}\left(k+1\right)=\left({X}^{\left(0\right)}\left(1\right)-b/a\right)\left(1-{e}^{-a}\right){e}^{a\left(k\right)}+\frac{b}{a},k=1,2,\cdots ,n$$
(4)

Where, X(0) is the original data series, \({\hat{X}}^{\left(0\right)}\) the predicted value, a is the development coefficient, and b is the gray action, which is used to describe the change trend and level of the original data.

Results

Evolution characteristics of Chinese dietary structure

During the 30-year period from 1992 to 2021, the Chinese dietary structure evolved from a single-type diet to a diversified diet, and dietary nutrient intake was gradually enriched and in surplus (Fig. 2). Specifically, the diversity index of the Chinese dietary structure increased from 0.46 in 1992 to 0.84 in 2021, reflecting a shift from homogenization to greater diversification. Additionally, regional variations in dietary structure were observed. The characteristics of food consumption have shifted from being primarily grain-based to incorporating more animal foods and a greater variety of food types (Fig. 2a, b, e). In terms of nutrient intake levels, the Chinese dietary structure has evolved from “satisfying satiety” to “eating nutritiously” (Fig. 2c, d). Dietary caloric intake decreased from 2436.1 kcal in 1992 to 2187.4 kcal in 2021. Protein intake fell from 90.3 g to 84.9 g, while fat intake increased from 50.7 g to 84.5 g, primarily due to a rise in high-quality protein consumption from animal foods. Dietary fiber intake decreased from 98.8 g to 81.3 g, and carbohydrate intake decreased from 392.8 g to 274.5 g.

Fig. 2: Changes of Chinese dietary structure (1992–2021).
figure 2

a Food consumption in 1992. b Food consumption in 2021. c Diet nutrient in 1992. d Diet nutrient in 2021. e Dietary diversity index from 1992 to 2021. f Comparison of Chinese dietary structure and Dietary Guidelines 2022. *In this paper, all maps are produced according to the Chinese standard map-Audit No GS(2019)1822, with no alterations to the base map.

Full size image

Compared to the balanced dietary pattern recommended by the Chinese Dietary Guidelines 2022, the Chinese dietary structure exhibits limited variety in grain intake, insufficient vegetable and fruit consumption, and excessive intake of livestock and meat (Fig. 2f). Caloric intake exceeded the balanced dietary pattern by 185.66 kcal, protein intake by 4.36 g, fat intake by 3.29 g, and carbohydrates by 33.04 g. Dietary fiber intake was 23.03 g below the recommended level. The increased consumption of animal foods has led to protein and fat intakes surpassing the dietary guidelines’ recommendations. This indicates that, although the Chinese dietary structure is evolving towards greater nutritional adequacy, it remains skewed towards a diet rich in nutrients.

Evolution characteristics of healthcare expenditure in China

From 1992 to 2021, China’s national health expenditure increased rapidly, with the expenditure structure exhibiting two increases and one decrease (Fig. 3). Over the past 30 years, China’s total health expenditure has consistently risen, accompanied by ongoing adjustments in its expenditure structure. China’s total health expenditure rose from 109.686 billion yuan in 1992 to 7684.499 billion yuan in 2021, representing an increase of approximately 69.06 times. Government health expenditure grew from 22.861 billion yuan to 2067.606 billion yuan. Its proportion of total health expenditure decreased from 20.8% in 1992 to 18.1% in 2007, then increased to 27.4%. Social health expenditure increased from 43.155 billion yuan to 3496.326 billion yuan. Its proportion initially fell from 39.3% to 38.1% in 2015, then rose to 44.3%. Personal cash health expenditure increased from 43.67 billion yuan to 2.2120.567 billion yuan, and its proportion increased from 39.8% to 60.0% in 2002 and then decreased to 28.4%. Per capita healthcare costs rose from 50.29 yuan to 2115.00 yuan. State financial expenditure on healthcare increased from 5838 million yuan to 1,914,268 million yuan, with its proportion of total state financial expenditure rising from 1.56% to 7.79%.

Fig. 3: Evolution of Healthcare Expenditure in China (1992–2021).
figure 3

a Changes in China’s health care resource investment. b China’s provincial health expenditure structure in 2021.

Full size image

The structure of healthcare expenditures by province shows a highly balanced structure in the east and a less balanced structure in the west. The elasticity of China’s total health expenditure growth relative to GDP has remained high, around 2, over the past 30 years. China’s total health investment remains low compared to the WHO’s target of at least 5% of GDP for government health spending. Statistics for 2020 show that China has reached only 5.6%, Japan 10.9%, the UK 12.0%, the US 18.8%, and the world average 10.9%. China’s health spending is below the global average and falls short compared to developed countries, where health investment typically exceeds 10% of GDP and has not yet reached saturation.

Effect of healthcare expenditure in China

Table 1 presents the results of a fixed-effects model analyzing various food consumption patterns and healthcare expenditures. The results showed that on a per capita basis, a significant association exists between grains, fruits, eggs, fish, and dairy consumption, including per capita and household healthcare expenditure. Conversely, oilseeds, vegetables, and meat were associated with lower per capita and household healthcare expenditures. At the government level, grain, vegetables, fruits, eggs, and dairy were associated with lower local financial health expenditures, whereas oilseeds, fish, and meat were significantly associated with increased local financial health expenditures. Additionally, grains, oilseeds had a significant impact on lower personal cash expenditure and total health costs, while fruits, vegetables, and eggs were significantly associated with increased personal cash expenditure and total health costs. Among the remaining control variables, mortality, aging, and disposable income have a more significant impact on healthcare expenditures. For individual residents, plant-based diets predominantly had a significant effect on increased healthcare expenditures, whereas oilseeds and meats predominantly had an effect on lower healthcare expenditures. However, at the government level, plant-based diets were significantly associated with lower healthcare expenditures, whereas animal-based diets were associated with increased healthcare expenditures.

Table 1 Fixed effect model of food consumption and health expenditure.
Full size table

Table 2 presents the results from a fixed-effects model analyzing the relationship between nutrient intake and healthcare expenditures. The results indicated that calorie and dietary fiber were significantly associated with lower per capita healthcare expenditures. Protein, fat, and carbohydrate significantly increased per capita healthcare expenditures. Moreover, both calorie and dietary fiber were significantly associated with lower household healthcare expenditures. Protein, fat, and carbohydrate significantly increased household healthcare expenditures. Additionally, calorie and dietary fiber significantly increased local fiscal health expenditures, whereas protein, fat, and carbohydrate significantly reduced fiscal health expenditures. Furthermore, calorie and dietary fiber were significantly associated with lower personal cash expenditures and total health costs, while protein, fat, and carbohydrate intake were significantly associated with increased expenditures for both categories. Among the control variables, the aging population and disposable income were associated with increased per capita and household healthcare expenditures, whereas the urbanization rate was associated with lower healthcare expenditures. The number of health checks, health institutions, beds, and the urbanization rate significantly increased government health expenditures, while the number of health technicians and the death rate significantly decreased expenditures. Moreover, the number of health checks significantly reduced personal and total healthcare expenditures, whereas the number of health institutions significantly increased these expenditures.

Table 2 Fixed effect model of various food nutrition elements and health expenditure.
Full size table

Forecast of China’s healthcare expenditure in 2030

The Chinese dietary structure is expected to evolve towards a high-protein, high-fat model in the future (Table 3). Specifically, grain consumption is projected to decline, reaching 139.2 kg in 2025, 144.11 kg in 2030, and 144.63 kg in 2035. Grain consumption will briefly increase in 2040 and 2045 before declining to 138.48 kg by 2050. Vegetable consumption shows a general upward trend, reaching 113.06 kg by 2030 and 122.76 kg by 2050. Fruit consumption remains relatively stable, peaking at 60.85 kg by 2030 and slightly declining to 59.02 kg by 2050. Oilseeds are on an upward trend, reaching 11.78 kg in 2030 and 15.1 kg in 2050. Eggs and dairy show minimal change. The fish category will rise only to 17.48 kg in 2050. Meat consumption will increase significantly in 2035, then stabilize above 50 kg. In terms of nutrient intake, food calorie intake is on an upward trend, reaching 2350.03 kcal by 2050, and protein intake is also gradually rising, reaching a peak value of 93.24 g by 2045 and then dropping to 88.87 g by 2050. Fat intake will increase sharply, from 84.99 g in 2025 to 100.43 g by 2050. Carbohydrate intake is projected to rise, reaching 217.67 g by 2050, while dietary fiber intake will increase to 90.87 g by 2050.

Table 3 Food consumption and nutrient intake of residents in China from 2025 to 2050.
Full size table

We estimated future changes in foodborne healthcare expenditures using data on the predicted dietary nutrient intake of the Chinese population (Fig. 4). According to the projected diet-healthcare system model, by 2030, foodborne per capita healthcare expenditure in China is expected to decrease by 41.66 yuan, while household, state financial, personal, and total healthcare expenditures are expected to rise by 76.83 yuan, 18.76 billion yuan, 95.28 billion yuan, and 17.69 billion yuan, respectively. By 2050, the estimated coefficients predict changes in healthcare expenditures due to food nutrient intake as follows: per capita foodborne expenditure will decrease by 2217.79 yuan, household expenditure by 2535.20 yuan, personal expenditure by 512.20 billion yuan, and total expenditure by 484.46 billion yuan. National financial health expenditure is projected to increase by 182.13 billion yuan.

Fig. 4: Forecast of Chinese dietary structure and changes of food-borne medical expenditure.
figure 4

a The Chinese dietary structure in 2030, 2050. b Predictive analysis of Chinese dietary structure and dietary recommendations. c The predicted value of China’s medical expenditures scale from 2022 to 2030.

Full size image

Additionally, using the gray prediction model, we forecasted healthcare expenditures assuming no major shocks or reforms to the system. By 2030, per capita, national financial, personal, and total healthcare expenditures in China are projected to reach 2563.98 yuan, 2547.14 billion yuan, 28948.32 billion yuan, and 10441.56 billion yuan, respectively. Compared to the baseline year of 2021, per capita, household, and government health expenditures are expected to increase. However, this increase surpasses the impact of changes in foodborne healthcare expenditures due to dietary shifts. This result suggests that while changes in dietary structure affect healthcare expenditure, future economic development and advances in medical technology also significantly influence healthcare costs.

Discussion

Exploring the relationship between the evolution of the Chinese dietary structure and healthcare expenditures is a further deconstruction of the complex relationship between preventive-end inputs and curative expenditures. This result highlights that the adjustment of the dietary structure is a priority for optimizing the structure of the national healthcare expenditures.

Relationship between dietary structure and health

The health sector plays a crucial role in enhancing national health. Diet is a fundamental component of national health, and the effects of dietary changes on health have been widely recognized. Over a quarter of global deaths are linked to eating disorders (GBD, 2019), primarily due to diet-related chronic diseases (Muka et al., 2015; Springmann et al., 2016). Adopting a nutritionally balanced diet is recommended as a key strategy to enhance nutritional status (Springmann et al., 2018a; Springmann et al., 2018b; Willett et al., 2019). Recent studies have also focused on the health effects of air pollution levels correlated with dietary modifications (Springmann et al., 2023). The study suggested that China’s great burden of disease (i.e., 40,000 premature deaths in three years) can be attributed to dietary risks, including low intake of fruits, nuts, and roughage and high intake of oil and salt (GBD, 2019; World Cancer Research Fund, 2018). Our projections predict that future dietary patterns in China will shift towards reduced grain consumption and increased intake of oils and meats. This case will lead to an increase in the burden of disease, and thus, the healthcare expenditures of the Chinese will continue to increase, particularly in the area of personal healthcare expenditures.

Relationship between dietary structural health effect and medical expenditure

The impact of diet on health care costs has been extensively documented. Studies indicate that increasing cereal consumption could reduce type 2 diabetes (T2D) related costs by between €286 million and €989 million (Martikainen et al., 2021). Our findings reveal that plant-based foods, such as grains and vegetables, contribute significantly to health care savings, whereas animal-based foods, including meat and oil, are generally linked to higher health care costs. This aligns with Jardim et al. (2019), who report that high consumption of red meat and saturated fats is associated with increased healthcare costs. Additionally, our results indicate that higher dietary fiber intake (−22.17) and lower calorie intake (−279.0) are linked to reduced healthcare expenditures. This correlation has been confirmed by studies, such as those by Fayet-Moore et al. (2018), which analyze the link between increased dietary fiber intake and healthcare cost savings. Our projections suggest that future dietary changes in China could account for 12.3% of the overall shift in personal cash expenditures, highlighting the significant influence of diet on personal healthcare costs. However, changes in food-related healthcare spending represent only 0.2% of the total shift in healthcare costs. Consequently, government and social health expenditures have lessened the effect of dietary adjustments on individual cash expenditures.

The concept of food-as-medicine treatment has been demonstrated in various countries, such as the food as Medicine program in the United States (Sharma and Sharma, 2024). Additionally, a clinical trial examining the impact of eating behavior interventions on medical costs found that dietary changes can significantly reduce patients’ medical expenses (Kopp et al., 2024). FAO estimates indicated that a healthy diet can significantly lower medical costs (FAO, 2020). In the future, the impact of multiple complex dietary patterns on changes in healthcare spending systems needs to be deeply considered. In the context of China’s healthcare system development and reform, transitioning to a plant-based diet will be crucial for reducing personal costs and optimizing medical expenditure by increasing vegetable and fruit intake while decreasing meat consumption. Promoting dietary changes can be achieved through national dietary guidelines that also address environmental impact and medical costs. Additionally, research indicated that calorie labeling on food packaging can effectively support dietary interventions (Feucht and Zander, 2017), while public advertising campaigns and smart apps are also effective in promoting healthy eating (Weichselbaum et al., 2013; Biasini et al., 2021; Türkmenoğlu et al., 2021). All these measures can be applied to improve the Chinese diet.

China’s medical policy guarantees and supports

China’s healthcare policy has been pivotal in safeguarding and supporting the health of its citizens. Over the past 30 years, China’s healthcare security system has undergone significant reforms, which can be categorized into four stages: exploration, initial formation, continuous improvement, and deepening reform (Fig. 5). During the initial exploratory phase, healthcare funding transitioned from being solely dependent on government finances to a shared investment model involving both the government and hospitals. The medical security system evolved from fully reimbursed public healthcare in urban areas to a basic medical insurance system for urban workers and residents, primarily focusing on major illness coverage. Additionally, a new cooperative healthcare model was established in rural areas, integrating individual contributions, collective support, and government subsidies. In the continuous improvement stage, the healthcare security system combined the New Rural Cooperative Program and the Urban Resident Insurance Program into a unified medical insurance system for both urban and rural residents. Additionally, the healthcare security system facilitated the convergence of medical assistance and major disease insurance and proposed the concept of instant settlement for basic medical insurance when seeking treatment elsewhere.

Fig. 5
figure 5

Reform of China’s medical security system.

Full size image

During the deepening reform phase, a multi-tiered medical insurance system has been developed, consisting of supplementary, primary, and basic levels. This system is supported by health insurance, basic medical insurance, and financial assistance. The payment method for this system is multifaceted and composite, with a particular emphasis on payment for each type of disease. Over the past 30 years, China has established the world’s largest universal basic medical insurance network, encompassing over 1.35 billion participants. The coverage rate of this network has stabilized to more than 95%. Within the healthcare system, China continuously aims to reduce personal cash expenditure. The multilevel medical insurance system has played a crucial role in improving national health. However, the development of the dietary structure has brought about an increased risk of disease burden. This case has led to a notable rise in healthcare expenses, particularly personal healthcare expenses, thereby affecting China’s healthcare insurance system. In the future, China should not only enhance its medical security system and basic medical services but also promote health management services, intensify preventive healthcare and health education, and increase investment in disease prevention. This includes regular physical exams, chronic disease prevention and control, and incorporating dietary intervention programs for specific patients into healthcare plans. Such measures will help foster a positive cycle between medical expenditure and health outcomes.

Research limitations

Existing research has estimated the costs related to individual food or nutrient intake and medical expenses. However, most studies focus on estimating the medical burden on individuals and assessing the national burden at the population level. This study’s strength lies in using macro-scale dietary consumption data to examine how changes in dietary structure impact macro-scale healthcare expenditures. The analysis considers multiple perspectives, including food consumption and nutrient intake at national, local, household, and individual levels. Predictions about future changes in food consumption and healthcare expenditures provide a reference for defining future foodborne healthcare expenditure boundaries. However, population healthcare expenditures will vary with time, economic size, consumption levels, and national financial decisions.

First, we examined the dietary structure based on 13 common foods as defined by the Food and Agriculture Organization. However, data on sugars were too limited to be included in the analysis. Our analysis focused exclusively on the five major nutritional elements, which are primarily obtained from dietary intake. However, this approach did not account for other micronutrients, such as iron, zinc, and calcium, which significantly impact human health. Additionally, we did not consider the potential health impacts of additives used in food processing.

Second, national healthcare expenditure is influenced by several factors. Our study considers various factors, including demographic structure, residents’ income, air pollution, urbanization level, and the number of healthcare institutions, beds, and technicians. However, national healthcare expenditure undergoes complex and dynamic changes. Health emergencies or natural disasters can cause drastic changes in expenditure that our model cannot predict. Additionally, the gray forecasting model we used assumes an exponential growth of healthcare expenditures without accounting for changes in the elasticity of the healthcare system.

Furthermore, we have not yet investigated whether a threshold effect exists between health expenditure and diet. In the future, the relationship between different patterns of dietary structure and medical expenditure still needs to be further explored. We hope to include regional diet culture, residents’ health status, economic effect, resources and environment, and other factors in the calculation of our “diet-healthcare system.”

Conclusion

We believe that structural medical issues related to food sources are a global concern. Additionally, the adverse effects of dietary changes must be carefully considered in population health management and healthcare policies. On a macro scale, plant-based diets generally decrease health expenditures, while animal-based diets tend to increase them. When calorie and dietary fiber intake decreases, the intake of protein, fat, and carbohydrates increases, leading to higher medical expenditures, especially personal cash expenditure. Moreover, this increase is several times greater than other health expenditures. In the future, with China’s population shifting toward a high-protein, high-fat dietary structure, food-borne personal medical cash expenditures will increase by 95.28 billion yuan in 2030. This case illustrates that proactive investment in dietary adjustments can yield dual benefits, namely, improved health and reduced curative expenditures. More broadly, our study enhances understanding of the socio-economic impacts of dietary structure as the presence of quality, healthy human capital is fundamental for sustainable development.